Irrespective of the tier (screening step, tier 1 or higher tier), risk assessment for birds and mammals is performed by calculating the toxicity-exposure ratio (TER), which is given by the following equations:
Acute TER = LD50 / DDD
Reproductive (long-term) TER = NOAELrepro / DDD
Estimation of the Daily Dietary Dose (DDD) of the active substance in question is thus a key element in risk assessment. In higher tier risk assessment, as dealt with in the present document, the DDD is calculated for one or more real species (“focal species”) that are known to occur in the crop(s) in question. Calculation of the DDD shall as far as possible be based on diet compositions, which have actually been measured in the field (as opposed to the generic diets used at tier 1).
Basically the DDD is given by the following equation:
DDD = ((FIR x C x PD) / BW) x PT, where
FIR = Food intake rate of the focal species in question (g fresh weight per day)
C = Concentration of active substance in fresh diet (mg/kg)
PD = Fraction of a particular food type in diet
BW = Body weight of focal species (g)
PT = Fraction of diet obtained within treated area.
The food intake rate (FIR) depends on the daily energy expenditure (DEE) of the species, which is again related to the body weight. FIR (g) is calculated by dividing DEE (kJ) by the energy content in 1 g of diet.
The concentration C is directly available in the special case of treated seeds, but in all other cases C must be calculated from the residue per unit dose (RUD), application rate, number of applications1, half-life of compound etc. (cf. EFSA 2009).
For a mixed diet, (FIR x C x PD) must be calculated separately for each food type, and the resulting DDD is the sum of the contributions from each food type in diet.
In the remaining sections of this chapter, estimation of PD and PT and a few other issues of relevance for higher tier risk assessment and the use of this document are briefly discussed.
As mentioned in the introduction, a calculator tool (Excel spreadsheet) has been developed to facilitate the calculation of DDD and TER for each of the selected focal species. Please refer to the introductory page of the calculator tool for specific guidance on how to use this tool.
4.2Derivation of crop and growth stage specific PD values
In this document, species-specific diets (PD values) to be used in higher tier risk assessment have been defined for each relevant combination of focal species, crop, growth stage and time of the year. This is straightforward for single-diet species and also fairly easy for other species that occupy rather narrow food and/or habitat niches. It is more difficult, however, to specify crop and growth stage specific diets for omnivorous species which have a general occurrence in farmland, e.g. skylark Alauda arvensisand wood mouse Apodemus sylvaticus. This is mainly because the major published studies of diet (e.g. Green 1978 for skylark, Pelz 1989 for wood mouse) elucidate the diet in arable land in general, rather than in specific crops.
The following example illustrates the problem. Skylark diet in April is specified as follows by Green (1978):
However, cereal grain and monocotyledonous leaves are mainly available in cereal fields and therefore their share of the diet will probably be much smaller in, e.g., oilseed rape fields where grasses occurring as weeds are the only monocotyledons present and grain is only available as old spillage. The DDD of skylarks foraging in pesticide-treated rape fields will therefore be biased if it is estimated directly from the general PD data above.
Basically, two different (and mutually exclusive) approaches might be used to overcome this problem. One approach would be to assume that the diet in each month is fixed and that those food items which are not available in rape fields will be obtained from elsewhere. Thus PD is retained and a residue of zero is assigned to those food items which are assumed to be obtained outside the treated field. Calculation of DDD is rather straightforward and no PT factor shall be used (because foraging outside the treated area is already accounted for by assuming zero residues in some food items). This approach is not recommended. The other approach assumes that the animal adjusts its diet according to availability in the crop in question. Therefore PD is adjusted to reflect availability in, e.g., rape fields. This makes estimation of DDD less straightforward and the adjustment may introduce an element of subjectivity. However, this approach is more in line with the official definition of PD (“composition of diet obtained from treated area”, EFSA 2009) and is the approach used in the present context. Standard (or measured) RUD values are used for all food items occurring in the diet. Foraging outside the treated field can be accounted for by applying a PT factor.
Following this approach, the published PD values which apply to arable land in general were adjusted for each relevant combination of crop, growth stage and month, taking the relative availability of different food items in the crop and growth stage in question into account. Furthermore “invertebrates” were split into foliar and ground-dwelling arthropods and “vegetative plant tissue” was split into mono- and dicotyledonous plants because rather different RUD values apply to those groups (cf. the following section).
To deal with this problem in an objective way, a set of fixed criteria was developed and applied to the data. E.g. for skylark, in non-cereal crops the share of cereal grain in diet was reduced to 6 %, corresponding to the minimum level found by Green (1978)2, and the relative share of the other food items was increased proportionally.
The criteria used for skylark and wood mouse are specified in Appendix 1 and 2, respectively.
The major published studies of the diets of important focal species such as skylark (Green 1978) and wood mouse (Pelz 1989) rely on field data from the 1970s and 1980s. Agricultural conditions have changed profoundly since then and it is highly probable that the diets of key farmland species have changed as well, reflecting changes in food availability. New studies would therefore be welcome.
If new studies are to supersede the old ones, sample sizes must be adequate and should preferably be comparable to those of the old studies. Thus, dietary studies based on 10-20 fecal sacs/droppings, e.g. from animals caught for tagging, will not be accepted. Preferably, data should be peer-reviewed and available for inclusion in future revisions of the present document.